Qwen-Image-2.1 LoRA 合集 / LoRA Collection

基于 Qwen/Qwen-Image-2.1 训练的图像编辑 LoRA 合集,由 GooHAI 发布。 A collection of image-editing LoRAs fine-tuned on Qwen/Qwen-Image-2.1, released by GooHAI.

包含模型 / Included LoRAs

1. 一致性无偏移 GH_V1.0

文件 / File: Qwen2.1 一致性无偏移_GH_V1.0.safetensors

  • 用途 / Purpose:解决 Qwen2.1 像素偏移问题。
  • Purpose: Resolves the pixel offset issue of Qwen2.1.
  • 触发词 / Trigger:无(不需要触发词 / no trigger word)
  • 推荐权重 / Recommended scale:0.7

2. 万物移除 Remove_GH

文件 / File: Qwen2.1_万物移除Remove_GH.safetensors

  • 用途 / Purpose:精准移除遮罩指定区域,无像素偏移、无色差。
  • Purpose: Precisely removes the masked specified region, with no pixel offset and no color deviation.
  • 触发词 / Trigger:移除图像中的绿色区域
  • 推荐权重 / Recommended scale:1.0

使用方法 / Usage

  1. 加载基础模型 Qwen/Qwen-Image-2.1。

  2. 按推荐权重加载对应的 .safetensors LoRA。

  3. 配合触发词(如适用)生成或编辑图像。

  4. Load the base model Qwen/Qwen-Image-2.1.

  5. Load the corresponding .safetensors LoRA at the recommended scale above.

  6. Use the trigger word where applicable to generate or edit the image.

许可证 / License

Apache License 2.0 (apache-2.0).

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for GooHAI/Qwen-image-2.1_loras

Adapter
(167)
this model